Race and Ethnicity, Structural Racism and Timely Breast Biopsy

Abstract

Background: Minority race and ethnicity are often associated with delays in breast cancer diagnosis and treatment. Structural racism may contribute to such delays and may prove to be an important target for interventions to reduce disparities in healthcare delivery. Methods: We conducted a single institution retrospective cohort study of 1938 women,18 years old or older who received a screening mammogram and a subsequent recommendation for breast biopsy. We conducted unadjusted and adjusted logistic regression analyses to determine the relationship between race and ethnicity and nonreceipt of a breast biopsy within 30, 60, and 90 days. Adjusted models included woman-, neighborhood- and healthcare-level variables of interest, including neighborhood-level measures of structural racism such as historical redlining. We also sought to quantify the relative importance of these multilevel factors in predicting nonreceipt of biopsy using a variable important analysis. Results: In unadjusted logistic regression models, Black women had higher odds of not receiving a breast biopsy within 30, 60, or 90 days (OR [95% CI]; 30 days: 2.27 [1.63,3.15], 60 days: 2.53 [1.72, 3.73], 90 days: 2.16 [1.37, 3.42]). After adjustment for multilevel variables, Black women continued to have greater odds of not receiving a timely biopsy (OR [95% CI]; 30 days: 1.96 [1.35, 2.83], 60 days: 2.31 [1.35, 2.83], 90 days: 2.1 [1.50,3.55]. Neighborhood-level variables, such as historical redlining, were not associated with nonreceipt of biopsy in the multivariable regression. However, in an unadjusted logistic regression model of historical redlining on nonreceipt of biopsy within 30 days of abnormal screening, living in “C” grade neighborhoods as associated with greater odds of not receiving a biopsy compared to living in an “A” grade neighborhood (OR [95% CI]; 4.11 [1.18,14.28]). Further sensitivity analyses limited to women living within graded neighborhoods showed that Black women had no significant difference in odds of not receiving a biopsy compared to White women and historical redlining was significantly associated with receipt of biopsy within 60 and 90 days (p = 0.39 and p =0.45, respectively). Variable importance analyses did not identify any variables having importance measure estimates greater than 0 in accurately predicting nonreceipt of biopsy. Conclusion: Black women with a recommendation for breast biopsy had greater odds of not receiving a biopsy within 30, 60, or 90 days when compared to White women. While initial analyses failed to detect any neighborhood-level variables as significantly associated with nonreceipt of biopsy, neighborhood historical redlining was associated with nonreceipt of biopsy in an adjusted sensitivity analysis limited to women living in neighborhoods graded “A”- “D.” Future work including more patients living within neighborhoods assessed by the Home Owners’ Loan Corporation are needed to better characterize the role of neighborhood redlining on racial and ethnic disparities in timely delivery of breast biopsies after an abnormal screening mammogram.